JPS61229961A - Method and apparatus for controlling operation of internal combustion engine - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a method and apparatus for controlling the operation of an internal combustion engine, and more particularly, to generating control data values as a function of operating parameters of the internal combustion engine and controlling the operating parameters. at least one data generator for controlling the influencing engine variables;
The control data values from the redata generator are corrected by a closed-loop control device responsive to the actual values of two engine variables, and the control data values read out and stored in the data generator according to the operating parameters of the internal combustion engine are The present invention relates to a method and apparatus for controlling the operation of an internal combustion engine, which is varied and self-adjusted through a control device.

[Prior Art] German patent application P 3408215.9 by the applicant
In the method for controlling the operation of an internal combustion engine as described above, the stored data value, which is stored in a data generator and selected according to operating parameters such as the rotational speed of the internal combustion engine and the load, is changed according to learning and is simply set to a predetermined value. A method is described in which not only one control data value of , but also the control data values located around it are modulated according to changes in the associated data values. Specifically, while the internal combustion engine is actually operating,
The integral controller continuously corrects the values read out from the data generator.

In this case, the data generator is divided into a predetermined number of sampling points, the intermediate values of which are arranged to be calculated by linear interpolation.

Simultaneously with the correction of the values read from the data generator, the multiplicative deviation correction coefficients of the controller are averaged, and this average value corresponds when leaving the operating section defined around each sampling point. It is calculated at the sampling point. In this way, by changing or moving the sampling point, the control data values are matched to the values given by the controller, and the control values are self-learning in their entirety, preventing only a predetermined area from being learned. . In this way, in the finely divided control data values - control data values are generated that are only slightly or totally selected and therefore have no learning effect, which are stored in the data generator. This makes it possible to prevent the control data values from changing over time.

Furthermore, as described in German Patent Publication No. 2847021 and British Patent No. 20349308, it is known that a fuel mixture is formed by metering or metering the fuel, for example via a learning control device. It is being Such learning control devices are equipped with a data generator in which, for example, injection control data values are stored and are stored in a read/write memory (RAM) each time the internal combustion engine is started. Data values are now being transferred. By providing such a data generator,
It becomes possible to obtain control values that respond very quickly to data such as injection quantity, typically fuel supply, ignition point that can be adjusted to changing operating conditions of the internal combustion engine, and exhaust gas recirculation rate. . In this case, in order to provide a learning effect to the control device, each control data value must be corrected according to the operating parameters and written into the memory respectively.

A self-learning injection device or a control device for controlling other operating parameters is provided with a data generator storing data such as the injection time as input values, ie, address values, such as the rotational speed and the throttle valve position. This data generator is subdivided into idle, part load, full load and deceleration areas (engine braking). In the idling range, the rotational speed is feedback-controlled, and in the partial load range, the fuel consumption is minimized, and in the full load range, the output is maximized. Further, in a deceleration region (engine brake, etc.), fuel supply is cut off. In any case, in order to adapt the control data values to the respective values provided by the controller, a learning method is introduced, commonly referred to as self-adjustment of the control values. The controller mentioned above is
Any actual value of the controlled object can be processed, and the output signal output according to the deviation between the actual value and the target value multiplicatively corrects the value read from the data generator according to the address such as rotation speed or load. has been done. In this case, the control data area to be learned is acted upon via the average value of the deviation correction coefficients.

For example, if the object to be controlled is an internal combustion engine, the engine variables that are processed as actual values are the input sensor installed in the exhaust pipe,
is the output signal from an oxygen sensor or other suitable sensor,
In the case of extreme value control that controls minimum fuel consumption or maximum output, it is the rotational speed of the internal combustion engine, etc.

[Problems to be Solved by the Invention] However, such conventional methods and devices have the disadvantage that the adaptive control time is long and therefore a relatively long time is required to generate the learning effect, and the The problem was that it was not possible to quickly adapt the entire control data value over a range of .

Therefore, the present invention has been made to eliminate such conventional drawbacks, and an object of the present invention is to provide an internal combustion engine operation control method and apparatus that can match control data values at high speed. .

[Means for Solving the Problems] In order to solve these problems in the present invention,
When the control data value stored in the data generator is changed through the closed-loop control device, a predetermined proportion of the change is formed as a comprehensive coefficient, and the control data value obtained from the data generator is determined by the comprehensive coefficient. We adopted a configuration in which a multiplicative or additive action is applied to the data, and all data sampling points are displaced in a multiplicative or additive manner.

[Operation] With such a configuration, the data generator is accessed using operating parameters such as the rotational speed and load of the internal combustion engine as address values, and control data values such as injection time stored therein are read out. This control data value is corrected for deviation by a cascaded closed loop control circuit so as to approach the target value. In the seventh case, a self-adjustment or learning effect is imparted to the data values stored in the data generator via the average value of the deviation correction coefficients. Furthermore, a global coefficient is determined which acts multiplicatively on the overall control data value via the average value of the deviation correction coefficients. By introducing such a comprehensive coefficient, it becomes possible to speed up the adaptive control or self-adjustment function and provide an optimal learning effect.

[Example] Hereinafter, the present invention will be described in detail based on the example shown in the drawings.

There are two basic features of the present invention, one of which is that the self-adjusting data generator is divided into a basic data generator whose data value does not change and a coefficient value generator whose data value changes. The basic data value read out according to the respective input address and the coefficient stored at the same address read out from the coefficient value generator are multiplied together. Another feature of the invention is the provision of global coefficients which preferably act multiplicatively and/or additively on the overall control data values.

In addition, although the circuits that realize these features are illustrated as block diagrams or functional diagrams in the drawings, each element or block can be configured in an analog, digital, or hybrid manner, and can also be configured using a microprocessor. This can be realized through program-controlled digital devices such as microcomputers, digital logic circuits, etc.

Figure 1 shows an internal combustion engine such as an externally ignited gasoline engine, a self-ignited diesel internal combustion engine, a Wankel engine, a Stirling engine, or a gas turbine, each of which is operated intermittently or by a fuel injection device or any fuel supply means. A control device that combines open-loop control and closed-loop control of an internal combustion engine with continuous injection is schematically illustrated as a block diagram. Although the figure shows an example of controlling the fuel injection pulse te, the invention is not limited to such engine variables, but can also be applied to other variables, such as exhaust gas composition.

It can also be applied to control of rotation circle immersion degree 1 boiling timing, boost pressure, exhaust gas recirculation rate, etc. or idling control.

5U East 1 In FIG. 1, in order to quickly form the fuel injection amount, the system is divided into a region 10 for generating control data values (open loop control region) and a closed loop control region 11 connected in cascade thereto. With this closed-loop control, the control data values read out from the data generator 12 are multiplicatively acted upon at 13 via addresses defined according to operating parameters such as rotational speed and load. Since the controller 14 must operate anew at each operating point, the open loop control region 10
A self-regulating circuit 15 is provided for adaptive control or learning based on the output from the controller. This self-regulating circuit automatically adjusts the control data value at each operating point to m! H, the deviation from the controller 14 will rapidly decrease.

Through such self-adjustment, other control data values (operating sections) located around the changed control data value are modulated according to the change in each control data value, preferably with weights added, so that the control data It becomes possible to adapt the values precisely and quickly to the actual operating conditions of the internal combustion engine 16.

In order to quickly optimize the self-adjustment of control data values by compensating for additive and multiplicative disturbances, the self-adjustment circuit 15 of the present invention is configured as follows, as shown in FIG. . In addition, FIG. 2 shows a fuel injection system in which input control (air-fuel ratio feedback control US>, extreme value control, etc.) are connected in cascade as an example. First, self-learning of data values is as follows.

(a) As mentioned above, the injection time is determined by the basic data generator 20, preferably configured as a fixed memory (ROM).
By accessing, for example, the rotation speed n, negative load air amount QL, and throttle valve position α) as an address, the control data value (tK) corresponding to this address is read out according to the number of sampling points and the number of interpolation. formed by

(b) Self-adjustment (adaptive learning) with a separate coefficient value generator 21, preferably configured as a read/write memory (RAM).
It is done through. This coefficient value generator is accessed in parallel with the basic data generator 20 by the same address. For this purpose, the elementary data generator 20 is preferably divided into several regions with predetermined values, each region being assigned a value from the coefficient value generator. Within these regions, the output signal tK from the elementary data generator is multiplied (or added) at the point of action 22 by the coefficient F read out from the coefficient value generator.

(C) In this case, self-adjustment by the coefficient value generator takes place only when the operating point is in steady state.

(d) In FIG. 2, the second feature of the invention is also introduced, in which the basic data generator is A so-called global factor is used which acts multiplicatively (or additively) on all values from 20 onwards. In that case, the comprehensive coefficient is the deviation correction coefficient RF obtained from the controller 23.
or according to the value from the coefficient value generator 21 described above. In Figure 2, this comprehensive coefficient generator is 2
4, and the points of action by this comprehensive coefficient are 25
Illustrated in

Further, in FIG. 2, the controller 23 and the variable that drives this controller and indicates the actual value of the internal combustion engine that is the controlled object (λ value, or the rotation speed such as the rotation speed fluctuation in the extreme value control described above) are shown. A measuring device 26 for measuring is shown. Although the coefficient value generator and the comprehensive coefficient generator in FIG. 2 are shown in combination in FIG. 2, they can of course be used independently of each other or singly. In any case, in the example shown in FIG. 2, the final injection time ti is tistK e F e G F 6 RF. The global factor GF acts multiplicatively and/or additively on each control data value read from the data generator. On the other hand, the coefficient F obtained from the coefficient value generator 21
has a local effect, so the coefficient value generator is accessed with the same address value as the elementary data generator 20,
Driven in parallel. Further, in FIG. 2, an average value forming circuit 28 for forming an average value RF of the deviation correction coefficients RF from the controller 23 is provided. In this case, as described above, the comprehensive coefficients are each determined according to the average value RF of the deviation correction coefficients or the value from the coefficient value generator 21.

Hereinafter, with reference to FIG. 3, a method for calculating the self-vlim of data values and the comprehensive coefficient through correction using the comprehensive coefficient will be described. FIG. 3 shows the formation of the fuel injection value via a data generator and the cascade closed-loop control, which is exemplified as an extreme value control. In addition, in Fig. 3, Fig. 1,
Parts that are the same as in FIG. 2 are given the same reference numerals, and slight differences are marked with a dash. In FIG. 3, the amount of fuel supplied to the internal combustion engine 27 is determined by the number of revolutions n
, the throttle valve position DK (or angle α) is read out from the data generator 12 as an address value. The throttle valve 29 is driven by an accelerator pedal 30. Data generator 1
The injection time ti stored in 2 is transmitted through the injection valve 31,
It is converted into a corresponding amount of fuel QK. The amount of air QL determined by the amount of fuel and the position of the accelerator pedal is supplied to the internal combustion @rjJJ27, and torque M is obtained according to the λ value (air ratio) of the air-fuel mixture. The internal combustion engine 27 to be controlled is approximated as an integral function as illustrated by block 27a. Internal combustion engine output (
The rotational speed n) together with the throttle valve position is used as an input value for the data generator 12.

The pure open-loop control described above is cascaded with closed-loop control based on the principle of extreme value control (the actual value of closed-loop control is based on other internal combustion engine output values, such as exhaust gas composition and rotational immersion). In the case of extreme value control, the air amount QL is varied by a predetermined amplitude ΔQL via the bypass path, or the injection time ti is varied by an amplitude Δti. . The test signals required for this purpose are outputted by a test signal generator 32 and, depending on the type of extreme value control, act on the fuel or air quantity either at a constant or at a variable frequency dependent on the rotational speed. Air volume QL
Alternatively, torque fluctuations occur due to periodic fluctuations in the amount of fuel, and the torque fluctuations are transmitted to the measuring device 3 as revolution speed fluctuations.
Detected by 3. The measuring device 33 analyzes the rotational speed fluctuations and correlates the fluctuation frequency and its influence by amplitude or phase. A comparison circuit 34 is connected after the measuring device 33 and compares the setpoint value with the actual value, the output of which is input to a controller 35 . The controller 35 generates a deviation correction coefficient RF and corrects the value read from the data generator 12.

Preferably, the controller 35 is configured as an integrator, and an average value forming circuit 36 for averaging the deviation correction coefficients is connected at a subsequent stage. The output signal 100 of this average value forming circuit acts on the control data value or the sampling value of the data generator 12 via the switch St.
This is done in the form of a weighted reduction correction.

The switch Sl, 52.33 is driven by the output signal of the area detection circuit 37, which is driven in parallel by the input value or address value of the data generator 12. switch 5
2.53 resets the average value forming 7 circuit 36 and the controller 35 to their initial values. The region detection circuit 37 determines which region (for example, fitting,
Detect which operating section is defined by the sampling point distance (partial load, all-load and deceleration operation area) or the sampling point distance of the station, and accordingly calculate the average value RF of each of the deviation correction coefficients by sampling the data generator 12. It will be counted as a point. Further, the comprehensive coefficient forming circuit 39 is activated via the lead wire 38, and the controller 35 and the average value forming circuit 36 are simultaneously reset to their initial values.

In the embodiment shown in FIG. 3, the output signal GF from the global coefficient forming circuit 39 and the deviation correction coefficient R from the controller 35 are
Rather than acting separately on the control data values of the data generator 12, F is directed to a common multiplication or summing point 40 such that each control value te from the data generator is applied to the multiplication point 4.
1, it is applied multiplicatively (or additively) and a total correction is performed. In the embodiment of FIG. 3, the comprehensive coefficient G
F is determined from the average value of the deviation correction coefficients, which will be explained in detail below.

The first method for determining the comprehensive coefficient GF is as follows.

When changing the control data value, it is possible to check how much the data value has changed. A predetermined percentage of the amount of change is taken into the inclusive layer number GF.

Since each control data value obtained from the data generator or each interpolated control data value is multiplied (added) by this global coefficient via multiplication points 40, 41, this global coefficient multiplies all sampling points. It has the function of displacing it in an additive manner.

In the embodiment of FIG. 3, the controller 35 forms a deviation correction factor RF from the control deviation so that the control data values interpolated from the data generator continue to be multiplied via the multiplication points 40, 41. , or acted upon additively. When the engine speed or the throttle valve position changes and leaves the operating zone defined by the previous sampling points, the average value of the deviation correction coefficient RF is incorporated into the data generator, and S
The control value of the data generator is changed according to the formula 5Sn=SSn11RF, where Sn is the new sampling value and SSa is the previous sampling value.

At the same time, a part of this correction is formed as a global coefficient GF via a global coefficient forming circuit 39 composed of a microprocessor or the like. In that case, this global factor is formed approximately according to the formula GFn=GFa+a* (RF-1), where a is the action factor, GFa is the previous and GFn is the new comprehensive factor.

This comprehensive coefficient has an integral characteristic with a large time constant. Since the comprehensive coefficient is changed only when adjusting the control data value, a wide range of data can be referenced to determine the comprehensive coefficient. This number of inclusive layers and deviation correction coefficient are
A multiplicative (additive) effect is applied at the point of action illustrated at 40 in FIG. Corrections are made.

Generally, the control data value can be changed to the target value in a multiplicative manner, i.e., by a multiplicative action that primarily causes the overall change in the data value to occur, but also acts additively on the overall control data value. Alternatively, the target value can be changed by an action that changes the structure of the data value. Experiments have shown that both effects can be performed separately, but optimal correction can be achieved by adjusting the sampling points as well as adjusting the coverage factor. In this case, the more complete the multiplicative correction of the control data value by the comprehensive coefficient is, the longer the transition time will be. Therefore, as a compromise solution, a preferable result can be obtained by performing the multiplicative correction using the comprehensive coefficient at a rate of about 50%, and performing the rest by changing the sampling point.

Introducing the coverage factor in this way allows better adjustment in addition to the adjustment by sampling point displacement.

When a car is stopped for a long time, the control data values may change considerably during this time due to changes in air pressure, temperature, etc., for example. After starting,
If such global changes occur in the data generator and the global coefficients have not yet been determined, it is possible that a correctly adjusted data structure may become incorrect. Therefore, in the present invention, the area detection circuit 37
After start-up via , the comprehensive coefficient is determined for a predetermined period of time, and the data generator is activated only when a new value of the comprehensive coefficient is formed. On the other hand, in order to avoid having to calculate the comprehensive coefficient again when the car is stopped for a short period of time, the comprehensive coefficient is activated only when the internal combustion engine has warmed up.

Another method for determining the comprehensive coefficient GF is as follows.

When adjusting control data values.

GF n = GF a%f (a, RF)
--- Predetermined percentage a of the deviation correction coefficient according to the formula (1)
is incorporated into the comprehensive coefficient. In that case, equation (1) is changed to 1/a
If repeated use occurs, the overall (averaged) deviation correction coefficient should be obtained.

f (a, RF) 1/a = RF or f
(a, RF) = RF a'' (2) That is, the comprehensive factor is multiplied by RF a at each adjustment.

GF n = GF a-) IRF a---(
3) The control data values obtained from the data generator are multiplied by a new global coefficient based on an interpolation method.

Manipulated amount=SS≠(RF bowl GF) However, SS is the control data value or sampling value read from the data generator.

In order to avoid variations in control data values, total deviation correction factors shall not be integrated into the data generator, SS
a. Set RFa to the previous control data value, the average value of the deviation correction coefficient, and set SSn and RFn to new values, respectively, and set SSa -NRFa %GFa-9Sn %RFn %
GFIl=SSn≠1%(GFa-NRFa a)(
3) According to the formula, SSa舛RFa*SSn Juice RFa aSSnmS
Sa arrowRFa / RFa aSSnmS
Sa %RFa (1-a)...(4)(3)
In order to reduce the production cost with respect to

GFn=GFa+an(RF-1)"15) (with respect to 0, the action coefficient raJ is 5 usually chosen to be very small, i.e. a<<1. Therefore, compared to 1, it can be ignored, and as mentioned above As above, we obtain S S n = S S a Homare RF-(8).

Additionally, experiments have shown that when the above-described calculations are performed, the same proportion of correction of control data values is only partially detected in the comprehensive coefficients. This is because as long as the global coefficient has not yet reached its final value, its components are introduced into the data generator.

Figures 4 to 7 show the final values and transient characteristics of the comprehensive coefficients (the effect coefficients are different in Figure 7) obtained from measurements and experiments, and the comprehensive coefficients and control data values. It can be seen that the changes are uniform. For this purpose, the actual data values (corresponding to the control data values of the control device 0), the setpoint data values (corresponding to the ideal values for the engine), the circulating signal generator (the driving signal generated by the driver) are used. (corresponding to the characteristics) are determined, and the learning methods (5) and (8) described above are used as the basis thereof.

Testing is performed using computer simulation. In that case, the uniform distribution of the correction is not affected in any way and the test of the control data group can be reduced to the test of one data characteristic. The circular signal generator generates the address of the actual sampling point of the data generator, the quotient of the target sampling point and the actual sampling point is used as the actual correction coefficient, and the comprehensive coefficient and control data group are determined according to the learning method described above, respectively. distributed to.

In that case, the simulation is carried out until the system is stable, i.e. until the global coefficients do not change. By changing parameters such as a large value deviation, its structure, and the type of cyclic signal (ie, whether it is random at the sequence end), the characteristics shown in FIGS. 4 to 7 can be obtained. FIG. 4 shows the part incorporated into the comprehensive coefficient of uniform deviation, normalized to the total deviation of the target data value, and 6 action coefficients a shown for the action coefficient a. is illustrated logarithmically. The characteristic curve (■) shown in FIG. 4 is related to eight operation sampling points, where GF=GF+a (positive 1-1) and correction amount=RF%GF.
The characteristic curve ■ is related to 16 operating sampling points under the same conditions, and the characteristic curve ■ is an approximation when the deviation is 20% and there is no multiplication or sI calculation, and the characteristic curve ■ is related to the case where the deviation is 1001. It is something.

The respective characteristics in FIGS. 5, 6, and 7 show two simulation examples. Each figure is a characteristic curve when sequentially passing through sampling points 1 to 8, and the sampling values and comprehensive coefficient values in the circulation from SS1 to SS8 are illustrated.

As shown in Figures 5 and 6, 1 action coefficient is thick and a
= 0.5, most of the change is accounted for by the inclusion factor (final value is 80% after 20 passes), but the system stabilizes slowly (a = 0.5
After to passes in the case corresponds to 4 passes with a = 0.0825), and.

The transient characteristics are large.

The final value of the comprehensive factor is determined by the following calculations, each differing according to the effect factor. Let E be the final value of the coverage factor and SSA be the number of operating sampling points.

(a) E = f (a 5SJA ) The final value is related to the product of the effect factor and the motion sampling point. That is,
Even if the effect coefficient is doubled and the number of sampling points is halved, the same final value will be obtained. Of course, such a dependence is satisfied only for the straight part of the characteristic curve of FIG. 4 (for a final value of 50%, the turning point).

(b) E=0 for a=0 (c) E=0 for a=17 SSA.
5(d) For a=1 E=1-1/SSA (permanent oscillation) The largest final value that can be reached is related to the number of motion sampling points. For 5SA-8, it is 87.5% of the uniform change, and for 5SA-18 it is 93.75%, and for 5SA-
20 is 95%, etc.

(e) If the number of SS is infinite, E=1, and (f) Let SSK be the number of sampling points to be corrected.

E = f (SSK / 5SA).

The final value depends on the ratio of the sampling points to be corrected to the total number of operating sampling points. That is, if only one quarter of the operating sampling points are corrected, the global coefficient will also be one quarter of the final value.

Generally, if the correction is different for each sampling point,
Find the average value of all correction amounts and calculate the final value of the comprehensive coefficient. (g) E = f (4/n -W-ΣKi ) (h
) The final value is independent of the form of circulation.

Also, the transition periods are generally different.

That is, if you want to cycle from SSI to SS8 and back in the same sequence as from SSI, then from SSI to SS8 and then SSI
The transition period is shorter than in the case of sequentially cycling alternately from S8 to SSI.

When the address is accessed by a quasi-random signal generator, if the effect factor is large (&>3'l),
The transient period will be shorter than if the action factor were smaller.

When the comprehensive coefficient is calculated multiplicatively according to equation (3), the comprehensive coefficient is as follows.

GFn=GFae+RF' Also, the final value is smaller than the additive calculation according to formula (5), E mult C55HEF) = E add
(SSe+EF/1.4) The characteristics of the final value are (E=
0.5), which corresponds to the characteristics of additive calculations. Also, the transition periods are approximately the same.

When applied to automobiles, a method that does not use multiplication and division is preferable due to calculation time considerations; in this case, the control value interpolated by the data generator is not further multiplied by the global coefficient, but rather the deviation The correction factor as well as the global factor are added to the interpolated control data value before multiplication.

That is.

Operation amount = SS hakama (RF + GF) Then, adjust the control data value as follows.

SSa 4 (RF + CF)-5Sn (1+
CF) SSn=SSa N[(RF+CF) / (1
+CF month Therefore, a division is required to calculate the new sampling point. Such a complicated calculation is (8
) to multiply the deviation correction coefficient and comprehensive coefficient,
It can be approximated as follows.

5Sn=SSa-NRF When performing additive calculations, the final value is generally related to the amount of sampling correction required. If the correction is large and the effect coefficient is large, the comprehensive coefficient will have a much larger value than the characteristic curve shown in FIG. (Refer to characteristic curves ■ and ■) When displacing 100% control data value, a÷0.
From the effect coefficient of 14, the inclusive coefficient becomes a negative value. others,
The transition time becomes noticeably longer.

In such a method, the coefficient of action is a
It is better not to choose a value larger than = 0.1.

4 Self-adjustment of the coefficient value generator FIG. 8 schematically shows the principle of self-adjustment of the control data values of the data generator. The field of control data values is divided into a basic data generator 20, preferably implemented in a fixed memory (ROM), and a coefficient value generator 21. The basic data generator 20 stores corresponding data at sampling points, the intermediate values of which are calculated by linear interpolation. The number of sampling points and the number of interpolated intermediate values are determined according to the required quantization for the corresponding control method. For fuel injectors, the quantization is such that the data generator has 1flX1B sampling points, each with 1
It is configured as a data generator with 5 intermediate values.

Self-adjustment is carried out using a further coefficient value generator 21 of Is2. This is preferably read/write memory (RAM)
The self-adjustment value is stored there. The basic data generator is subdivided into a plurality of areas, and each area corresponds to a coefficient from the coefficient value generator 21. The interpolated output values from the basic data generator 20 are each multiplied by an associated coefficient or a value interpolated from a plurality of coefficients, and multiplied (or added) at a multiplication point (summing point) 22. In the illustrated embodiment, 8×8 coefficients are provided, each output value having a value of rl, OJ, which undergoes a corresponding change during the adjustment process.

The final injection value is determined by the basic control data value tK read out from the basic data generator, the coefficient F obtained from the coefficient value generator 21 and the deviation correction factor RF obtained from the closed-loop control circuit, and optionally It is obtained by multiplying by other correction coefficients.

In other words, if written in one equation, it becomes titan tK・F-RF.If the operating point is changed to a region with another coefficient F of the coefficient value generator, a fluctuation will occur in the output value, but if this is a fault, In such a case, it can be prevented by setting the deviation correction coefficient RF accordingly.

Preferably, each coefficient F of the coefficient value generator 21 is interpolated;
How does the impact of such interpolation affect learning?
VKMo will explain how it affects you. The adjustment of the hole coefficient obtained from the coefficient value generator 21 is given by the following formula: However, FA is the previous coefficient and FN is the new coefficient. Therefore, as long as a certain region of the elementary data generator 20 is selected, the deviation correction factors RF are averaged and, via block 40, the factor F associated therewith is varied.

FIG. 9 shows the structure of a basic data generator 20 with 18×1B sampling points.

From this data generator, the fuel injection pulse ti is determined according to the throttle valve position DK (=Y) and the rotation speed n (=X).
The value of is read. In the data group illustrated in Figure 9,
There are shaded areas and non-shaded areas, and each shaded area (total of 64 areas) indicates an operation section, and for each operation section, the coefficient value generator 21 Common coefficients are stored in the coefficient value generator 21. As already mentioned, the coefficient value generator has 8.times.8 coefficients, and the operating section illustrated in FIG. 9 can be divided arbitrarily.

Adjustment of the coefficients is performed as illustrated in FIG. FIG. 10(a) shows the output from the basic data generator 20, in which the driving characteristics are entered. Also shown is the operating section C for the selected coefficients. A indicates the entry point of the running characteristics in this operating section, and B indicates the exit point of this running curve.

FIG. 10(b) shows the characteristic of the deviation correction coefficient RF versus time. After entering the operating zone illustrated in FIG. 1O(a), a predetermined transient delay occurs. Subsequently, the deviation correction coefficients are averaged. In this case, Fig. 10(b)
Try to get the minimum average time as shown in the diagram. B
At a point, after leaving the operating section or after the minimum averaging time has elapsed, the average value RF of the deviation correction coefficients is calculated according to the above-mentioned formula and is included in the constant F.

The above-mentioned transient times as well as minimum average times make it possible to distinguish between steady-state and dynamic operating points. As already mentioned, it is best to make adjustments only under steady-state conditions, and also during warm-up, after start-up,
The system is designed to shut off during fuel cut and acceleration enrichment. The functions performed by the area detection circuit 37 shown in FIG. 3 can also be realized using, for example, a computer controlled according to its program.

By using the coefficient value generator 2x*m, any misalignment of the elementary data generator 20 can be corrected by using a suitable closed-loop control. In this case, all corrections are performed in a steady state, except for areas that are not often selected. Therefore, disturbances that act additively or multiplicatively can be optimally corrected by using a coefficient value generator, and in the case of multiplicative disturbance components that act uniformly, by forming comprehensive coefficients. , it becomes possible to make corrections.

The following table shows a list of disturbances that act in a multiplicative or additive manner, as well as their effects when used in the Alpha M system (a system in which the input signals that form the injection time are the throttle valve position and rotation speed). The characteristics are illustrated. The times at which these disturbance amounts change are different.

FIG. 11 shows in detail the method for determining the coverage coefficients described above. In this case, the first step to find the comprehensive coefficient is
In this method, the average value of the deviation correction coefficients determined by the average value forming circuit 28' is inputted via the switch S4.
Via the attenuators 41 and 42, the coefficient value generator 21 shown in Fig. WS8 as well as the read/write memory (similar to the coefficient value generator)
RAM). It is input to the global coefficient generator 24'. The averaging of the deviation correction factors RF is carried out as long as the operating point is in a predetermined operating section of the basic data generator 20.

After a predetermined period of time has elapsed or when leaving the operating area,
An adjustment of the corresponding coefficient F takes place as already mentioned, and the global coefficient GF is changed only when this operating section changes. A new coefficient FA from the coefficient value generator as well as a new global coefficient G F pa are formed according to the formulas described below.

As is clear from each formula, a part of the average value of the deviation is included in the related coefficient, and another part is included in the comprehensive coefficient.

A method for determining comprehensive coefficients according to the method illustrated in FIG. 11 is illustrated as a flowchart in FIG. In step 31, the address N of the coefficient value generator is calculated, and when it is determined in step S2 that the address has changed, it is determined in step S3 whether or not the average time has elapsed. When the average time has elapsed, a correction coefficient F and a comprehensive coefficient GF are calculated in step S4. After setting the marker flag in step S5, branching is performed in step S5. In step S6, the address N of the correction coefficient F is determined.

The average value RF of the deviation correction coefficient, the average time counter X, and the transient time counter Y are set to initial values.

If the transient time has elapsed in step S7, the deviation correction coefficients are averaged in step S8 and the average time counter is incremented, and then the transient time counter is incremented in step S9.

If this method is the first method, another method for determining the global coefficient, ie, the second method (2), is shown as a block diagram in FIG. 12 and as a flowchart in FIG. 14, 151g.

In the block diagram shown in FIG. 12, a second coefficient value generator 21'' is also provided, which data generator includes a basic data generator 20 as well as a first coefficient value generator 21'.
In parallel, the same input data (namely rotational speed and load) are input as addresses and likewise act multiplicatively on the basic data values. That is, the first multiplication point 43 and the second multiplication point 44
At this multiplication point 44, which acts via the base data generator 20, a total correction factor acts on the injection data value te read out from the basic data generator 20. The coefficient value generator 21' is set to 1.0 when the internal combustion engine is started and is continuously adjusted. The data of the coefficient value generator 21' as well as the global coefficient remain unchanged; furthermore, the marker generator determines whether this coefficient is to be selected.

After a predetermined time has elapsed, the coefficient value generator 21'' is processed. The deviation between the initial value 1.0 and the average value of all coefficients is included in the global coefficient (via switch 46 and lead 45). , while the remaining structural deviation of 1.0 is incorporated into the coefficient value generator 21'.

However, only selected coefficients are taken into account. Thereafter, the coefficient value generator is again set to a value of 1.0 and the adjustment is made again in a similar manner. According to this second method, calculation is performed using a linear equation when determining a comprehensive coefficient.

GFN −GF A ” 1/n ΣFLz
GFA +(1/n Σ(FL-1))+
1 Also, the following equation can be obtained from the changed sampling point F.

Fz-PL/ (1/n Σ'c) k Fz-
(1/n Σ(FL-1)) The corresponding program implementing the method of this formula consists of two parts. The first part corresponds to the first method illustrated in FIG.
In this case, no comprehensive factor is calculated (b=o)
, the second part is yet another subprogram of the first method, and as shown in FIG. 14, corresponding locations are marked with circles. In addition, a second method for determining comprehensive coefficients is performed in the software domain, and R
AM can also be omitted and all calculations can be performed using only the coefficient value generator 21'. A control flow illustrating this method is illustrated at 151M.

In FIG. 14, in step 510, a counter AZ that determines the time during which the correction coefficient F is adjusted is incremented, and when it is determined in step 511 that the adjustment time has ended, in step 512, all of the coefficient value generators 21'' are average the coefficients, adjust the comprehensive coefficient, and send the changed coefficient value to the coefficient value generator 2
Incorporate into 1'. Subsequently, the AZ counter is reset, and FIL and all marker flags are reset.

In FIG. 15, step 520 increments AZ;
After the adjustment time has elapsed in step 521, all coefficients are averaged and optionally amplified in step 322. If it is determined in step S23 that the comprehensive coefficient needs to be adjusted, in step 524, the comprehensive coefficient is set to al[L, the changed coefficient is reset at tR, and all coefficients are averaged. If it is determined in step S25 that a displacement of the control data value is necessary, the displacement is performed in step 526, and the GF is changed accordingly. Subsequently, in step S27, the AZ counter is reset and all marker flags MF are reset.

[Effects of the Invention] As described above, according to the present invention, disturbances that act multiplicatively or additively on necessary portions of control data values can be reduced to the entire control data value by using a so-called comprehensive coefficient. can be adjusted and quickly aligned, which provides significant speed compared to methods that adjust side-by-side control data values or sampling points. Further, according to the present invention, control data areas that are rarely or rarely selected can be adjusted quickly and accurately.

Furthermore, in the present invention, by dividing the data into a basic data group and a coefficient value data group subjected to self-adjustment, it is possible to prevent interpolation performed in the basic data area from having any negative effect on the learning method. In particular, the additive effect is caused by the self-adjusting coefficient value data (coefficient value generator),
In addition, disturbances can be taken into account, and among the disturbances, multiplicative effects that are components that act uniformly can be compensated for by combining the above-mentioned comprehensive coefficients and dispersions.

Additive as well as multiplicative effects allow fast and optimal adjustment.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a schematic configuration of the present invention, FIG. 2 is a block diagram explaining a first embodiment of the present invention, FIG. 3 is a block diagram showing a detailed embodiment for calculating comprehensive coefficients, and FIG. 4
The figure is a characteristic diagram showing the final value of the comprehensive coefficient for the action coefficient,
Figures 5(a)-(d). Figures 6(a) to (d) are characteristic diagrams showing the transient characteristics of the comprehensive coefficient for a given action coefficient, and Figures 7(a) to (d) are similar to Figure 6 for different action coefficients. FIG. 8 is a block diagram showing the second embodiment of the present invention. FIG. 9 is a three-dimensional diagram showing the characteristics of injection time as a function of throttle valve position and rotation speed. Explanatory diagram, 1st
FIG. 10(a) is a characteristic diagram showing entry and departure of the basic data generator, FIG. 10(b) is a characteristic diagram showing the flow of the deviation correction coefficient over time, and FIG. 11 is another embodiment of the present invention. FIG. 12 is a block diagram showing still another embodiment of the present invention, and FIGS. 13 to 15 are flowcharts showing the flow of control, respectively. 12...Data generator 14...Controller 20...
Basic data generator 21... Coefficient value generator 24... Comprehensive coefficient generator 2
8... Average value forming circuit 26... Measuring device 27... Internal combustion engine Fi9.
15

Claims (1)

Claims: 1) a data generator for generating control data values as a function of operating parameters of the internal combustion engine for controlling engine variables that influence the operating parameters, the actual value of at least one engine variable; The control data values from the data generator are corrected by a closed-loop controller responsive to the
A method for controlling the operation of an internal combustion engine, in which control data values read out and stored in a data generator according to operating parameters of the internal combustion engine are changed and self-adjusted via the closed-loop control device, when the control data values are changed; A method for controlling the operation of an internal combustion engine, characterized in that a predetermined ratio of is formed as a comprehensive coefficient, and the comprehensive coefficient applies a multiplicative or additive effect to a control data value obtained from a data generator. . 2) The global coefficient obtained by averaging the deviation correction coefficients and the actual deviation correction coefficient are integrated into one correction coefficient, which acts multiplicatively or additively on the control data value obtained from the data generator. An internal combustion engine operation control method according to claim 1. 3) The engine variable is the exhaust gas composition, rotation smoothness, rotation speed, etc. of the internal combustion engine, and the control data value from the data generator and the self-adjustment amount of the control data value are adjusted by the deviation correction coefficient. The method for controlling the operation of an internal combustion engine according to item 1 or 2. 4) The internal combustion engine operation control method according to any one of claims 1 to 3, wherein the internal combustion engine is a gasoline or diesel internal combustion engine, a Wankel engine, a Stirling engine, or a gas turbine. 5) Any one of claims 1 to 4, wherein the control is used for fuel-air mixture formation, ignition timing control, boost pressure control, exhaust gas recirculation rate control, and idling control. A method for controlling the operation of an internal combustion engine as described. 6) a closed-loop control device responsive to the actual value of at least one engine variable, comprising a data generator for generating control data values as a function of operating parameters of the internal combustion engine and for controlling engine variables that influence the operating parameters; The control data value from the data generator is corrected by
A method for controlling the operation of an internal combustion engine, in which control data values read out and stored in a data generator according to operating parameters of the internal combustion engine are varied and self-adjusted via said closed-loop control device, the deviation obtained by the closed-loop control device Operation control of an internal combustion engine, characterized in that a predetermined ratio of correction coefficients is used to form a comprehensive coefficient, and the comprehensive coefficient is used to apply a multiplicative or additive effect to a control data value obtained from a data generator. Method. 7) The engine variables include the exhaust gas composition, rotation smoothness, rotation speed, etc. of the internal combustion engine, and the control data value from the redata generator and the self-adjustment amount of the control data value are adjusted by the deviation correction coefficient. The method for controlling the operation of an internal combustion engine according to scope 6. 8) The internal combustion engine operation control method according to claim 6 or 7, wherein the internal combustion engine is a gasoline or diesel internal combustion engine, a Wankel engine, a Stirling engine, or a gas turbine. 9) Any one of claims 6 to 8, wherein the control is used for fuel-air mixture formation, ignition timing control, boost pressure control, exhaust gas recirculation rate control, and idling control. A method for controlling the operation of an internal combustion engine as described. 10) A closed-loop control device responsive to the actual value of at least one engine variable, comprising a data generator for generating control data values as a function of operating parameters of the internal combustion engine and for controlling engine variables that influence the operating parameters. The control data values from the data generator are corrected by the internal combustion engine, and the control data values read out and stored in the data generator are varied via the closed-loop control device according to the operating parameters of the internal combustion engine, so that the self-regulating internal combustion In an engine operation control method, in order to self-adjust control data values, the data generator is divided into a basic data generator formed by a fixed memory and a coefficient value generator for correction, and 1. A method for controlling the operation of an internal combustion engine, characterized in that a multiplicative or additive effect is applied to a predetermined area of a basic data generator using a coefficient. 11) The control data value obtained by the basic data generator is multiplicatively or additively affected by the comprehensive coefficient,
11. The method for controlling the operation of an internal combustion engine according to claim 10, wherein each correction is performed by a multiplicative or additive action using a coefficient obtained from a coefficient value generator in accordance with an operating parameter of the internal combustion engine. 12) The engine variables include the exhaust gas composition, rotation smoothness, rotation speed, etc. of the internal combustion engine, and the control data value from the redata generator and the self-adjustment amount of the control data value are adjusted by the deviation correction coefficient. The method for controlling the operation of an internal combustion engine according to item 10 or 11. 13) The amount of disturbance that mainly acts multiplicatively is determined by the comprehensive coefficient,
The method for controlling the operation of an internal combustion engine according to any one of claims 10 to 12, wherein the amount of disturbance acting additively is corrected by a coefficient from a coefficient value generator. 14) In order to determine the global coefficient as well as the coefficient from the coefficient value generator, the deviation correction coefficients are averaged while the internal combustion engine remains in each operating section of the basic data generator, and the deviation correction is performed when leaving the operating section. 11. The internal combustion engine operation control method according to claim 10, wherein each of the coefficients is changed by including a predetermined proportion of the average value of the coefficients. 15) A part of the average value of the deviation correction coefficient as a comprehensive coefficient,
15. The method for controlling the operation of an internal combustion engine according to claim 14, wherein a part of the coefficient is incorporated into the coefficients of the coefficient value generator. 16) Adjusting the coefficients of the coefficient value generator via the average value of the deviation correction coefficients and by defining the operating section in the basic data generator corresponding to the coefficient at a predetermined time or when leaving the operating section. 11. The method for controlling the operation of an internal combustion engine according to claim 10, wherein a predetermined proportion of the average value of the deviation correction coefficients is incorporated into a corresponding coefficient of a coefficient value generator. 17) The method for controlling the operation of an internal combustion engine according to claim 10, wherein the basic data generator is configured by a fixed memory and the coefficient value generator is configured by a random access memory. 18) After entering the predetermined operation zone and after a predetermined time elapses,
Claim 17, wherein the deviation correction coefficients are averaged, and then, after leaving the operating section or after a predetermined period of time has elapsed, the average value of the deviation correction coefficients is included in the coefficient value corresponding to the operating section. A method for controlling the operation of an internal combustion engine. 19) Furthermore, a second
Claim 1 provides a coefficient value generator for
The method for controlling the operation of an internal combustion engine according to item 0. 20) The internal combustion engine operation control method according to any one of claims 10 to 19, wherein the internal combustion engine is a gasoline or diesel internal combustion engine, a Wankel engine, a Stirling engine, or a gas turbine. 21) Any one of claims 10 to 20, wherein the control is used for fuel-air mixture formation, ignition timing control, boost pressure control, exhaust gas recirculation rate control, and idling control. A method for controlling the operation of an internal combustion engine as described. 22) A closed-loop control device responsive to the actual value of at least one engine variable, comprising a data generator for generating control data values as a function of operating parameters of the internal combustion engine and for controlling engine variables that influence the operating parameters. The control data values from the data generator are corrected and the control data values read out and stored in the data generator are varied and self-adjusted via the closed-loop control device according to the operating parameters of the internal combustion engine. In an operation control device for an internal combustion engine, a controller (23,
35, 23'), an average value forming device (28, 36, 28') that averages the deviation correction coefficients from the controller, and a device (24, 24, 28') that forms a comprehensive coefficient based on the average value of the deviation correction coefficients.
39), and the data generator (
All control data values (te, 20) obtained from
1. An operation control device for an internal combustion engine, characterized in that a multiplicative or additive effect is applied to tK). 23) A closed-loop control device responsive to the actual value of at least one engine variable, comprising a data generator for generating control data values as a function of operating parameters of the internal combustion engine and for controlling engine variables that influence the operating parameters. The control data values from the data generator are corrected and the control data values read out and stored in the data generator are varied via the closed-loop control device in accordance with the operating parameters of the internal combustion engine to provide a self-regulating internal combustion engine. In the engine operation control device, a controller (23, 3
5, 23'), an average value forming device (28, 36, 28') that averages the deviation correction coefficients from the controller, and a device (24, 24, 23') that forms a comprehensive coefficient based on the average value of the deviation correction coefficients.
39), dividing the data generator into a basic data generator and a coefficient value generator, correcting the coefficient (F) from the coefficient value generator based on the average value of the deviation correction coefficient, and An operation control device for an internal combustion engine, characterized in that control data values from a basic data generator are corrected in a multiplicative or additive manner. 24) The global coefficients and the coefficients from the coefficient value generator are
Claim 23 wherein the basic data values are corrected and self-adjusted based on the coefficient values.
The operation control device for an internal combustion engine according to paragraph 1. 25) Provide a second coefficient value generator into which the average value of the deviation correction coefficients is input, form a comprehensive coefficient based on the deviation of the average value of the coefficients of the first coefficient value generator, and calculate the remaining coefficient from the initial value. 25. The operation control device for an internal combustion engine according to claim 23, wherein the deviation is incorporated into the coefficients of the first coefficient value generator.